Failed to replace a bad datanode on the existing pipeline due to no more good datanodes being avail

错误：java.io.IOException: Failed to replace a bad datanode on the existing pipeline due to no more good datanodes being available to try.

2014-05-07 12:21:41,820 WARN [Thread-115] org.apache.hadoop.mapreduce.v2.app.MRAppMaster: Graceful stop failed
org.apache.hadoop.yarn.exceptions.YarnRuntimeException: java.io.IOException: Failed to replace a bad datanode on the existing pipeline due to no more good datanodes being available to try. (Nodes: current=[192.168.1.191:50010, 192.168.1.192:50010], original=[192.168.1.191:50010, 192.168.1.192:50010]). The current failed datanode replacement policy is DEFAULT, and a client may configure this via 'dfs.client.block.write.replace-datanode-on-failure.policy' in its configuration.
at org.apache.hadoop.mapreduce.jobhistory.JobHistoryEventHandler.handleEvent(JobHistoryEventHandler.java:514)
at org.apache.hadoop.mapreduce.jobhistory.JobHistoryEventHandler.serviceStop(JobHistoryEventHandler.java:332)
at org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
at org.apache.hadoop.service.ServiceOperations.stop(ServiceOperations.java:52)
at org.apache.hadoop.service.ServiceOperations.stopQuietly(ServiceOperations.java:80)
at org.apache.hadoop.service.CompositeService.stop(CompositeService.java:159)
at org.apache.hadoop.service.CompositeService.serviceStop(CompositeService.java:132)
at org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster.shutDownJob(MRAppMaster.java:548)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster$JobFinishEventHandler$1.run(MRAppMaster.java:599)
Caused by: java.io.IOException: Failed to replace a bad datanode on the existing pipeline due to no more good datanodes being available to try. (Nodes: current=[192.168.1.191:50010, 192.168.1.192:50010], original=[192.168.1.191:50010, 192.168.1.192:50010]). The current failed datanode replacement policy is DEFAULT, and a client may configure this via 'dfs.client.block.write.replace-datanode-on-failure.policy' in its configuration.
at org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.findNewDatanode(DFSOutputStream.java:860)
at org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.addDatanode2ExistingPipeline(DFSOutputStream.java:925)
at org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.setupPipelineForAppendOrRecovery(DFSOutputStream.java:1031)
at org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.processDatanodeError(DFSOutputStream.java:823)
at org.apache.hadoop.hdfs.DFSOutputStream$DataStreamer.run(DFSOutputStream.java:475)

原因：无法写入；我的环境中有3个datanode，备份数量设置的是3。在写操作时，它会在pipeline中写3个机器。默认replace-datanode-on-failure.policy是DEFAULT,如果系统中的datanode大于等于3，它会找另外一个datanode来拷贝。目前机器只有3台，因此只要一台datanode出问题，就一直无法写入成功。

解决办法：修改hdfs-site.xml文件，添加或者修改如下两项：
<property>
<name>dfs.client.block.write.replace-datanode-on-failure.enable</name>
<value>true</value>
</property>
<property>
<name>dfs.client.block.write.replace-datanode-on-failure.policy</name>
<value>NEVER</value>
</property>

对于dfs.client.block.write.replace-datanode-on-failure.enable，客户端在写失败的时候，是否使用更换策略，默认是true没有问题。
对于，dfs.client.block.write.replace-datanode-on-failure.policy，default在3个或以上备份的时候，是会尝试更换结点尝试写入datanode。而在两个备份的时候，不更换datanode，直接开始写。对于3个datanode的集群，只要一个节点没响应写入就会出问题，所以可以关掉。

转自：http://www.raincent.com/content-85-7554-7.html